scholarly journals In Situ Observation-Constrained Global Surface Soil Moisture Using Random Forest Model

2021 ◽  
Vol 13 (23) ◽  
pp. 4893
Author(s):  
Lijie Zhang ◽  
Yijian Zeng ◽  
Ruodan Zhuang ◽  
Brigitta Szabó ◽  
Salvatore Manfreda ◽  
...  
Keyword(s):  
Soil Moisture ◽  
Random Forest ◽  
Land Surface Temperature ◽  
Soil Texture ◽  
Land Surface ◽  
Surface Soil ◽  
Vegetation Indices ◽  
Surface Soil Moisture ◽  
Spatio Temporal

The inherent biases of different long-term gridded surface soil moisture (SSM) products, unconstrained by the in situ observations, implies different spatio-temporal patterns. In this study, the Random Forest (RF) model was trained to predict SSM from relevant land surface feature variables (i.e., land surface temperature, vegetation indices, soil texture, and geographical information) and precipitation, based on the in situ soil moisture data of the International Soil Moisture Network (ISMN.). The results of the RF model show an RMSE of 0.05 m3 m−3 and a correlation coefficient of 0.9. The calculated impurity-based feature importance indicates that the Antecedent Precipitation Index affects most of the predicted soil moisture. The geographical coordinates also significantly influence the prediction (i.e., RMSE was reduced to 0.03 m3 m−3 after considering geographical coordinates), followed by land surface temperature, vegetation indices, and soil texture. The spatio-temporal pattern of RF predicted SSM was compared with the European Space Agency Climate Change Initiative (ESA-CCI) soil moisture product, using both time-longitude and latitude diagrams. The results indicate that the RF SSM captures the spatial distribution and the daily, seasonal, and annual variabilities globally.

scholarly journals An evapotranspiration model self-calibrated from remotely sensed surface soil moisture, land surface temperature and vegetation cover fraction: application to disaggregated SMOS and MODIS data

2020 ◽  
Vol 24 (4) ◽  
pp. 1781-1803
Author(s):  
Bouchra Ait Hssaine ◽  
Olivier Merlin ◽  
Jamal Ezzahar ◽  
Nitu Ojha ◽  
Salah Er-Raki ◽  
...  
Keyword(s):  
Soil Moisture ◽  
Energy Balance ◽  
Land Surface Temperature ◽  
Satellite Data ◽  
Land Surface ◽  
Surface Soil ◽  
Heat Fluxes ◽  
Sensible Heat ◽  
Surface Soil Moisture

Abstract. Thermal-based two-source energy balance modeling is essential to estimate the land evapotranspiration (ET) in a wide range of spatial and temporal scales. However, the use of thermal-derived land surface temperature (LST) is not sufficient to simultaneously constrain both soil and vegetation flux components. Therefore, assumptions (about either soil or vegetation fluxes) are commonly required. To avoid such assumptions, an energy balance model, TSEB-SM, was recently developed by Ait Hssaine et al. (2018b) in order to consider the microwave-derived near-surface soil moisture (SM), in addition to the thermal-derived LST and vegetation cover fraction (fc) normally used. While TSEB-SM has been successfully tested using in situ measurements, this paper represents its first evaluation in real life using 1 km resolution satellite data, comprised of MODIS (MODerate resolution Imaging Spectroradiometer) for LST and fc data and 1 km resolution SM data disaggregated from SMOS (Soil Moisture and Ocean Salinity) observations. The approach is applied during a 4-year period (2014–2018) over a rainfed wheat field in the Tensift basin, central Morocco. The field used was seeded for the 2014–2015 (S1), 2016–2017 (S2) and 2017–2018 (S3) agricultural seasons, while it remained unploughed (as bare soil) during the 2015–2016 (B1) agricultural season. The classical TSEB model, which is driven only by LST and fc data, significantly overestimates latent heat fluxes (LE) and underestimates sensible heat fluxes (H) for the four seasons. The overall mean bias values are 119, 94, 128 and 181 W m−2 for LE and −104, −71, −128 and −181 W m−2 for H, for S1, S2, S3 and B1, respectively. Meanwhile, when using TSEB-SM (SM and LST combined data), these errors are significantly reduced, resulting in mean bias values estimated as 39, 4, 7 and 62 W m−2 for LE and −10, 24, 7, and −59 W m−2 for H, for S1, S2, S3 and B1, respectively. Consequently, this finding confirms again the robustness of the TSEB-SM in estimating latent/sensible heat fluxes at a large scale by using readily available satellite data. In addition, the TSEB-SM approach has the original feature to allow for calibration of its main parameters (soil resistance and Priestley–Taylor coefficient) from satellite data uniquely, without relying either on in situ measurements or on a priori parameter values.

scholarly journals Calibration of Noah Soil Hydraulic Property Parameters Using Surface Soil Moisture from SMOS and Basinwide In Situ Observations

2016 ◽  
Vol 17 (8) ◽  
pp. 2275-2292 ◽  
Author(s):  
Peter J. Shellito ◽  
Eric E. Small ◽  
Michael H. Cosh
Keyword(s):  
Soil Moisture ◽  
Soil Texture ◽  
Land Surface ◽  
Surface Soil ◽  
Soil Column ◽  
Surface Model ◽  
Surface Soil Moisture ◽  
Remotely Sensed Data ◽  
Soil Hydraulic

Abstract Soil hydraulic properties (SHPs) control infiltration and redistribution of moisture in a soil column. The Noah land surface model (LSM) default simulation selects SHPs according to a location’s mapped soil texture class. SHPs are instead estimated at seven sites in North America through calibration. A single-objective algorithm minimizes the root-mean-square difference (RMSD) between simulated surface soil moisture and observations from 1) a dense network of in situ probes, 2) Soil Moisture Ocean Salinity (SMOS) satellite retrievals, and 3) SMOS retrievals adjusted such that their mean equals that of the in situ network. Parameters are optimized in 2012 and validated in 2013 against the in situ network. RMSD and unbiased RMSD (ubRMSD) assess resulting surface soil moisture behavior. At all sites, assigning SHP parameters from a different soil texture than the one that is mapped decreases the RMSD by an average of 0.029 cm3 cm−3. Similar improvements result from calibrating parameters using in situ network data (0.031 cm3 cm−3). Calibrations using remotely sensed data show comparable success (0.029 cm3 cm−3) if the SMOS product has no bias. Calibrated simulations are superior to texture-based simulations in their ability to decrease ubRMSD at times of year when the default simulation is worst. Changes to both RMSD and ubRMSD are small when the default simulation is already good. Most calibrated simulations have higher runoff ratios than do texture-based simulations, a change that warrants further evaluation. Overall, parameter selection using SMOS data shows good potential where biases are low.

scholarly journals An evapotranspiration model self-calibrated from remotely sensed surface soil moisture, land surface temperature and vegetation cover fraction: application to disaggregated SMOS and MODIS data

2019 ◽  
Author(s):  
Bouchra Ait Hssaine ◽  
Olivier Merlin ◽  
Jamal Ezzahar ◽  
Nitu Ojha ◽  
Salah Er-raki ◽  
...  
Keyword(s):  
Soil Moisture ◽  
Energy Balance ◽  
Surface Temperature ◽  
Land Surface Temperature ◽  
Vegetation Cover ◽  
Land Surface ◽  
Surface Soil ◽  
Surface Soil Moisture ◽  
Wide Range ◽  
The Mean

Abstract. Thermal-based two-source energy balance modeling is very useful for estimating the land evapotranspiration (ET) at a wide range of spatial and temporal scales. However, the land surface temperature (LST) is not sufficient for constraining simultaneously both soil and vegetation flux components in such a way that assumptions (on either the soil or the vegetation fluxes) are commonly required. To avoid such assumptions, a new energy balance model (TSEB-SM) was recently developed in Ait Hssaine et al. (2018a) to integrate the microwave-derived near-surface soil moisture (SM), in addition to the thermal-derived LST and vegetation cover fraction (fc). Whereas, TSEB-SM has been recently tested using in-situ measurements, the objective of this paper is to evaluate the performance of TSEB-SM in real-life using 1 km resolution MODIS (Moderate resolution imaging spectroradiometer) LST and fc data and the 1 km resolution SM data disaggregated from SMOS (Soil Moisture and Ocean Salinity) observations by using DisPATCh. The approach is applied during a four-year period (2014–2018) over a rainfed wheat field in the Tensift basin, central Morocco, during a four-year period (2014–2018). The field was seeded for the 2014–2015 (S1), 2016–2017 (S2) and 2017–2018 (S3) agricultural season, while it was not ploughed (remained as bare soil) during the 2015–2016 (B1) agricultural season. The mean retrieved values of (arss, brss) calculated for the entire study period using satellite data are (7.32, 4.58). The daily calibrated αPT ranges between 0 and 1.38 for both S1 and S2. Its temporal variability is mainly attributed to the rainfall distribution along the agricultural season. For S3, the daily retrieved αPT remains at a mostly constant value (∼ 0.7) throughout the study period, because of the lack of clear sky disaggregated SM and LST observations during this season. Compared to eddy covariance measurements, TSEB driven only by LST and fc data significantly overestimates latent heat fluxes for the four seasons. The overall mean bias values are 119, 94, 128 and 181 W/m2 for S1, S2, S3 and B1 respectively. In contrast, these errors are much reduced when using TSEB-SM (SM and LST combined data) with the mean bias values estimated as 39, 4, 7 and 62 W/m2 for S1, S2, S3 and B1 respectively.

scholarly journals Accuracy and Transferability of Artificial Neural Networks in Predicting in Situ Root-Zone Soil Moisture for Various Regions across the Globe

Water ◽  
2020 ◽  
Vol 12 (11) ◽  
pp. 3109
Author(s):  
Roïya Souissi ◽  
Ahmad Al Bitar ◽  
Mehrez Zribi
Keyword(s):  
Neural Network ◽  
Soil Moisture ◽  
Soil Texture ◽  
Surface Soil ◽  
Root Zone ◽  
Data Filtering ◽  
Surface Soil Moisture ◽  
The Mean ◽  
Artificial Neural

This paper explores the accuracy in using an artificial neural network (ANN) to estimate root-zone soil moisture (RZSM) at multiple worldwide locations using only in situ surface soil moisture (SSM) as a training dataset. The paper also addresses the transferability of the trained ANN across climatic and soil texture conditions. Data from the International Soil Moisture Network (ISMN) were collected for several networks with variable soil texture and climate classes. Several scaling, feature extraction, and training approaches were tested. An artificial neural network employing rolling averages (ANNRAV) of SSM over 10, 30, and 90 days was developed. The results show that applying a standard scaling (SSCA) to the ANN input features improves the correlation, Nash–Sutcliffe efficiency (NSE), and root mean square error (RMSE) for 52%, 91%, and 87%, respectively, of the tested stations, compared to MinMax scaling (MMSCA). Different training sets are suggested, namely, training on data from all networks, data from one network, or data of all networks excluding one. Based on these trainings, new transferability (TranI) and contribution (ContI) indices are defined. The results show that one network cannot provide the best prediction accuracy if used alone to train the ANN. They also show that the removal of the less contributing networks enhances performance. For example, elimination of the densest network (SCAN) from the training enhances the mean correlation by 20.5% and the mean NSE by 42.5%. This motivates the implementation of a data filtering technique based on the ANN’s performance. A median, max, and min correlation of 0.77, 0.96, and 0.65, respectively, are obtained by the model after data filtering. The performances are also analyzed with respect to the covered climatic regions and soil texture, providing insights into the robustness and limitations of the approach, namely, the need for complementary information in highly evaporative regions. In fact, the ANN using only SSM to predict RZSM has low performance when decoupling between the surface and root zones is observed. The application of ANN to obtain spatialized RZSM will require integrating remote sensing-based surface soil moisture in the future.

scholarly journals Controls on surface soil drying rates observed by SMAP and simulated by the Noah land surface model

2017 ◽  
Author(s):  
Peter J. Shellito ◽  
Eric E. Small
Keyword(s):  
Soil Moisture ◽  
Soil Texture ◽  
Vegetation Cover ◽  
Land Surface ◽  
Surface Soil ◽  
Surface Model ◽  
Soil Drying ◽  
Potential Evaporation ◽  
Surface Soil Moisture ◽  
Drying Rates

Abstract. Drydown periods that follow precipitation events provide an opportunity to assess the mechanisms by which soil moisture dissipates from the land surface. We use SMAP (Soil Moisture Active Passive) observations and Noah simulations from drydown periods to quantify the role of soil moisture, potential evaporation, vegetation cover, and soil texture on soil drying rates. Rates are determined using finite differences over intervals of 1 to 3 days. In the Noah model, the drying rates are a good approximation of direct soil evaporation rates. Data cover the domain of the North American Land Data Assimilation System phase 2 and span the first 1.8 years of SMAP's operation. Drying of surface soil moisture observed by SMAP is faster than that simulated by Noah. SMAP drying is fastest when surface soil moisture levels are high, potential evaporation is high, and when vegetation cover is low. Soil texture plays a minor role in SMAP drying rates. Noah simulations show similar responses to soil moisture and potential evaporation, but vegetation has a minimal effect and soil texture has a much larger effect compared to SMAP. When drying rates are normalized by potential evaporation, SMAP observations and Noah simulations both show that increases in vegetation cover lead to decreases in evaporative efficiency from the surface soil. However, the magnitude of this effect simulated by Noah is much weaker than that determined from SMAP observations.

scholarly journals New Downscaling Approach Using ESA CCI SM Products for Obtaining High Resolution Surface Soil Moisture

2020 ◽  
Vol 12 (7) ◽  
pp. 1119 ◽  
Author(s):  
Jovan Kovačević ◽  
Željko Cvijetinović ◽  
Nikola Stančić ◽  
Nenad Brodić ◽  
Dragan Mihajlović
Keyword(s):  
Soil Moisture ◽  
High Resolution ◽  
Random Forest ◽  
Spatial Resolution ◽  
Surface Soil ◽  
Initial Step ◽  
Surface Soil Moisture ◽  
In Situ Data ◽  
Random Forest Models

ESA CCI SM products have provided remotely-sensed surface soil moisture (SSM) content with the best spatial and temporal coverage thus far, although its output spatial resolution of 25 km is too coarse for many regional and local applications. The downscaling methodology presented in this paper improves ESA CCI SM spatial resolution to 1 km using two-step approach. The first step is used as a data engineering tool and its output is used as an input for the Random forest model in the second step. In addition to improvements in terms of spatial resolution, the approach also considers the problem of data gaps. The filling of these gaps is the initial step of the procedure, which in the end produces a continuous product in both temporal and spatial domains. The methodology uses combined active and passive ESA CCI SM products in addition to in situ soil moisture observations and the set of auxiliary downscaling predictors. The research tested several variants of Random forest models to determine the best combination of ESA CCI SM products. The conclusion is that synergic use of all ESA CCI SM products together with the auxiliary datasets in the downscaling procedure provides better results than using just one type of ESA CCI SM product alone. The methodology was applied for obtaining SSM maps for the area of California, USA during 2016. The accuracy of tested models was validated using five-fold cross-validation against in situ data and the best variation of model achieved RMSE, R2 and MAE of 0.0518 m3/m3, 0.7312 and 0.0374 m3/m3, respectively. The methodology proved to be useful for generating high-resolution SSM products, although additional improvements are necessary.

Surface soil moisture data assimilation for irrigation amounts and timing estimation in semi-arid regions

2021 ◽  
Author(s):  
Nadia Ouaadi ◽  
Lionel Jarlan ◽  
Saïd Khabba ◽  
Jamal Ezzahar ◽  
Olivier Merlin
Keyword(s):  
Soil Moisture ◽  
Land Surface ◽  
Surface Soil ◽  
Land Surface Model ◽  
Surface Model ◽  
Surface Soil Moisture ◽  
Field Scale ◽  
In Situ Data ◽  
Statistical Metrics

<p>Irrigation is the largest consumer of water in the world, with more than 70% of the world's fresh water dedicated to agriculture. In this context, we developed and evaluated a new method to predict daily to seasonal irrigation timing and amounts at the field scale using surface soil moisture (SSM) data assimilated into a simple  land surface model through a particle filter technique. The method is first tested using in situ SSM before using SSM products retrieved from Sentinel-1. Data collected on different wheat fields grown  in Morocco, for both flood and drip irrigation techniques, are used to assess the performance of the proposed method. With in situ data, the results are good. Seasonal amounts are retrieved with R > 0.98, RMSE <42 mm and bias<2 mm. Likewise, a good agreement is observed at the daily scale for flood irrigation where more than 70% of the irrigation events are detected with a time difference from actual irrigation events shorter than 4 days, when assimilating SSM observation every 6 days to mimics Sentinel-1 revisit time. Over the drip irrigated fields, the statistical metrics are R = 0.70, RMSE =28.5 mm and bias= -0.24 mm for irrigation amounts cumulated over 15 days. The approach is then evaluated using SSM products derived from Sentinel-1 data; statistical metrics are R= 0.64, RMSE= 28.78 mm and bias = 1.99 mm for irrigation amounts cumulated over 15 days. In addition to irrigated fields, the applicationof the developed methodover rainfed fieldsdid not detect any irrigation. This study opens perspectives for the regional retrieval of irrigation amounts and timing at the field scale and for mapping irrigated/non irrigated areas.</p>

scholarly journals Cross-evaluation of modelled and remotely sensed surface soil moisture with in situ data in southwestern France

2010 ◽  
Vol 14 (11) ◽  
pp. 2177-2191 ◽  
Author(s):  
C. Albergel ◽  
J.-C. Calvet ◽  
P. de Rosnay ◽  
G. Balsamo ◽  
W. Wagner ◽  
...  
Keyword(s):  
Soil Moisture ◽  
Land Surface ◽  
Surface Soil ◽  
Root Zone ◽  
Weather Prediction ◽  
Remotely Sensed ◽  
Surface Model ◽  
Surface Soil Moisture ◽  
Moisture Profile

Abstract. The SMOSMANIA soil moisture network in Southwestern France is used to evaluate modelled and remotely sensed soil moisture products. The surface soil moisture (SSM) measured in situ at 5 cm permits to evaluate SSM from the SIM operational hydrometeorological model of Météo-France and to perform a cross-evaluation of the normalised SSM estimates derived from coarse-resolution (25 km) active microwave observations from the ASCAT scatterometer instrument (C-band, onboard METOP), issued by EUMETSAT and resampled to the Discrete Global Grid (DGG, 12.5 km gridspacing) by TU-Wien (Vienna University of Technology) over a two year period (2007–2008). A downscaled ASCAT product at one kilometre scale is evaluated as well, together with operational soil moisture products of two meteorological services, namely the ALADIN numerical weather prediction model (NWP) and the Integrated Forecasting System (IFS) analysis of Météo-France and ECMWF, respectively. In addition to the operational SSM analysis of ECMWF, a second analysis using a simplified extended Kalman filter and assimilating the ASCAT SSM estimates is tested. The ECMWF SSM estimates correlate better with the in situ observations than the Météo-France products. This may be due to the higher ability of the multi-layer land surface model used at ECMWF to represent the soil moisture profile. However, the SSM derived from SIM corresponds to a thin soil surface layer and presents good correlations with ASCAT SSM estimates for the very first centimetres of soil. At ECMWF, the use of a new data assimilation technique, which is able to use the ASCAT SSM, improves the SSM and the root-zone soil moisture analyses.

scholarly journals The impact of land surface temperature on soil moisture anomaly detection from passive microwave observations

2011 ◽  
Vol 15 (10) ◽  
pp. 3135-3151 ◽  
Author(s):  
R. M. Parinussa ◽  
T. R. H. Holmes ◽  
M. T. Yilmaz ◽  
W. T. Crow
Keyword(s):  
Soil Moisture ◽  
Surface Temperature ◽  
Land Surface Temperature ◽  
Land Surface ◽  
Surface Soil ◽  
Surface Soil Moisture ◽  
Ka Band ◽  
L Band ◽  
Evaluation Techniques ◽  
Soil Moisture Anomaly

Abstract. For several years passive microwave observations have been used to retrieve soil moisture from the Earth's surface. Low frequency observations have the most sensitivity to soil moisture, therefore the current Soil Moisture and Ocean Salinity (SMOS) and future Soil Moisture Active and Passive (SMAP) satellite missions observe the Earth's surface in the L-band frequency. In the past, several satellite sensors such as the Advanced Microwave Scanning Radiometer-EOS (AMSR-E) and WindSat have been used to retrieve surface soil moisture using multi-channel observations obtained at higher microwave frequencies. While AMSR-E and WindSat lack an L-band channel, they are able to leverage multi-channel microwave observations to estimate additional land surface parameters. In particular, the availability of Ka-band observations allows AMSR-E and WindSat to obtain coincident surface temperature estimates required for the retrieval of surface soil moisture. In contrast, SMOS and SMAP carry only a single frequency radiometer and therefore lack an instrument suited to estimate the physical temperature of the Earth. Instead, soil moisture algorithms from these new generation satellites rely on ancillary sources of surface temperature (e.g. re-analysis or near real time data from weather prediction centres). A consequence of relying on such ancillary data is the need for temporal and spatial interpolation, which may introduce uncertainties. Here, two newly-developed, large-scale soil moisture evaluation techniques, the triple collocation (TC) approach and the Rvalue data assimilation approach, are applied to quantify the global-scale impact of replacing Ka-band based surface temperature retrievals with Modern Era Retrospective-analysis for Research and Applications (MERRA) surface temperature output on the accuracy of WindSat and AMSR-E based surface soil moisture retrievals. Results demonstrate that under sparsely vegetated conditions, the use of MERRA land surface temperature instead of Ka-band radiometric land surface temperature leads to a relative decrease in skill (on average 9.7%) of soil moisture anomaly estimates. However the situation is reversed for highly vegetated conditions where soil moisture anomaly estimates show a relative increase in skill (on average 13.7%) when using MERRA land surface temperature. In addition, a pre-processing technique to shift phase of the modelled surface temperature is shown to generally enhance the value of MERRA surface temperature estimates for soil moisture retrieval. Finally, a very high correlation (R2 = 0.95) and consistency between the two evaluation techniques lends further credibility to the obtained results.

scholarly journals Evaluation of MERRA Land Surface Estimates in Preparation for the Soil Moisture Active Passive Mission

2011 ◽  
Vol 24 (15) ◽  
pp. 3797-3816 ◽  
Author(s):  
Yonghong Yi ◽  
John S. Kimball ◽  
Lucas A. Jones ◽  
Rolf H. Reichle ◽  
Kyle C. McDonald
Keyword(s):  
Soil Moisture ◽  
Land Surface ◽  
Surface Soil ◽  
Middle Latitude ◽  
Surface Soil Moisture ◽  
Freeze Thaw ◽  
Earth Observing System ◽  
In Situ Observations ◽  
Satellite Microwave

Abstract The authors evaluated several land surface variables from the Modern-Era Retrospective Analysis for Research and Applications (MERRA) product that are important for global ecological and hydrological studies, including daily maximum (Tmax) and minimum (Tmin) surface air temperatures, atmosphere vapor pressure deficit (VPD), incident solar radiation (SWrad), and surface soil moisture. The MERRA results were evaluated against in situ measurements, similar global products derived from satellite microwave [the Advanced Microwave Scanning Radiometer for Earth Observing System (EOS) (AMSR-E)] remote sensing and earlier generation atmospheric analysis [Goddard Earth Observing System version 4 (GEOS-4)] products. Relative to GEOS-4, MERRA is generally warmer (~0.5°C for Tmin and Tmax) and drier (~50 Pa for VPD) for low- and middle-latitude regions (<50°N) associated with reduced cloudiness and increased SWrad. MERRA and AMSR-E temperatures show relatively large differences (>3°C) in mountainous areas, tropical forest, and desert regions. Surface soil moisture estimates from MERRA (0–2-cm depth) and two AMSR-E products (~0–1-cm depth) are moderately correlated (R ~ 0.4) for middle-latitude regions with low to moderate vegetation biomass. The MERRA derived surface soil moisture also corresponds favorably with in situ observations (R = 0.53 ± 0.01, p < 0.001) in the midlatitudes, where its accuracy is directly proportional to the quality of MERRA precipitation. In the high latitudes, MERRA shows inconsistent soil moisture seasonal dynamics relative to in situ observations. The study’s results suggest that satellite microwave remote sensing may contribute to improved reanalysis accuracy where surface meteorological observations are sparse and in cold land regions subject to seasonal freeze–thaw transitions. The upcoming NASA Soil Moisture Active Passive (SMAP) mission is expected to improve MERRA-type reanalysis accuracy by providing accurate global mapping of freeze–thaw state and surface soil moisture with 2–3-day temporal fidelity and enhanced (≤9 km) spatial resolution.

Sign in / Sign up

Export Citation Format

Share Document